P80-Coilin: a Component of Coiled Bodies and Interchromatin Granule- Associated Zones
Total Page:16
File Type:pdf, Size:1020Kb
Journal of Cell Science 108, 1143-1153 (1995) 1143 Printed in Great Britain © The Company of Biologists Limited 1995 p80-coilin: a component of coiled bodies and interchromatin granule- associated zones Francine Puvion-Dutilleul1,*, Sylvie Besse1, Edward K. L. Chan2, Eng M. Tan2 and Edmond Puvion1 1Laboratoire de Biologie et Ultrastructure du Noyau de l’UPR 9044 CNRS, BP 8, F-94801 Villejuif Cedex, France 2The Scripps Research Institute, W. M. Keck Autoimmune Disease Center, 10666 N. Torrey Pines Road, La Jolla, California 92037, USA *Author for correspondence SUMMARY We investigated at the electron microscope level the fate of the clusters of interchromatin granules and their associated the three intranuclear structures known to accumulate zones, which were all easily recognizable within the snRNPs, and which correspond to the punctuate immuno- residual nuclear ribonucleoprotein network, was unmodi- fluorescent staining pattern (the coiled bodies, the clusters fied. The data indicate, therefore, that the loosening of interchromatin granules and the interchromatin procedure as well as the high salt extraction procedure granule-associated zones) after exposure to either a low salt preserve the snRNA content of all three spliceosome medium which induces a loosening and partial spreading component-accumulation sites and reveal that interchro- of nucleoprotein fibers or a high ionic strength salt medium matin granule-associated zones are elements of the nuclear and subsequent DNase I digestion, in order to obtain DNA- matrix. The p80-coilin content of coiled bodies was also depleted nuclear matrices. The loosened clusters of inter- preserved whatever the salt treatment used. An intriguing chromatin granules and the coiled bodies could no longer new finding is the detection of abundant p80-coilin within be distinguished from surrounding nucleoprotein fibers the interchromatin granule-associated zones, both before solely by their structure, but constituents of the clusters of and after either low or high salt treatment of cells. interchromatin granules could be detected by in situ Therefore, p80-coilin is an integral constituent of the inter- hybridization with both U1 and U2 DNA probes, and con- chromatin granule-associated zones. stituents of the coiled bodies were detectable mainly with the U2 DNA probe. The interchromatin granule-associated zones, the electron-opacity and compactness of which were Key words: coiled body, electron microscope, HeLa cell, in situ preserved despite the loosening treatment, remained hybridization, interchromatin granule, interchromatin granule- labeled with the U1 DNA probe only. In DNA-depleted associated zone, mild loosening procedure, nuclear matrix, p80- nuclear matrices, the snRNA content of the coiled bodies, coilin, spliceosome, U1 RNA, U2 RNA INTRODUCTION former and accumulate in considerable amounts in the latter. Both snRNAs accumulate within the clusters of interchro- The components of the cell nucleus involved in transcription matin granules (Spector, 1993; Visa et al., 1993a). As already and splicing have been clearly identified by immunocyto- mentioned (Visa et al., 1993a), the interchromatin granule- chemical and in situ hybridization procedures designed for associated zones, might be equivalent to the A snurposomes electron microscopy. It was clearly shown that the lower-defi- described in amphibian germinal vesicles by the Gall’s group nition speckled pattern observed by fluorescence microscopy (Gall, 1991; Wu et al., 1991). On the other hand, the very low following specific detection of spliceosome components level of labeling of the clusters of interchromatin granules on (Carmo-Fonseca et al., 1991, 1992; Huang and Spector, 1992; electron microscope autoradiographs following labeling of Lamond and Carmo-Fonseca, 1993a,b) corresponds to cells with radioactive uridine (Fakan, 1978; Fakan and perichromatin fibrils and to three different, well-defined struc- Puvion, 1980; Puvion and Moyne, 1981), as well as the total tures, i.e. the clusters of interchromatin granules, the inter- absence of DNA within these structures (Thiry, 1993; Visa et chromatin granule-associated zones and the coiled bodies al., 1993a), exclude the possibility that they could be sites of (Visa et al., 1993a). Although all three structures accumulate transcription and active splicing. Instead, it was proposed that spliceosome components, they differ in their specific con- they are more likely to be concerned with pre- and/or post- tents. U1 snRNAs are present in large amounts in the inter- splicing events such as the final maturation or storage of chromatin granule-associated zones but are much rarer in the snRNP particles, assembly of spliceosomes and/or intron coiled bodies. Conversely, U2 snRNAs are absent in the degradation (Lamond et al., 1990; Visa et al., 1993a,b). How- 1144 F. Puvion-Dutilleul and others ever, the permanent presence of poly(A)+ RNA in clusters of general organization of the clusters of interchromatin granules interchromatin granules (Visa et al., 1993b) and, at least in and their associated zones, whereas that of coiled bodies was certain metabolic conditions, of ribosomal RNA (Puvion et more profoundly changed. On the other hand, the high salt al., 1984b) suggests that they also could be sites of some sort- extraction of cells was without effect on the aspect (and ing and/or degradation of mature RNA molecules. Concerning snRNA content) of the three spliceosome component-accumu- coiled bodies, recent results have shown that this type of lation sites. An unexpected result is that anti-p80 coilin nuclear body could also be multifunctional and related to pre- antibody not only labeled coiled bodies of high salt-extracted, and/or post-splicing events and to unknown nucleolus-associ- loosened and unloosened materials, but also significantly ated functions (Brasch and Ochs, 1992; Lamond and Carmo- labeled the interchromatin granule-associated zones. Fonseca, 1993b; Malatesta et al., 1994). In fact, it is now gen- erally accepted that the perichromatin fibrils represent the MATERIALS AND METHODS main morphological substrate of splicing (Nash et al., 1975; Fakan et al., 1976; Puvion and Moyne, 1978; Puvion and Cell preparation Viron, 1981; Gallinaro et al., 1983). These fibrils form at tran- For electron microscope study, HeLa cells (5×105 cells per 5 cm scription sites, contain nascent RNA (Nash et al., 1975; Fakan plastic culture dish) were grown at 37°C, in the presence of 5% CO2, et al., 1976; Puvion and Moyne, 1978; Puvion and Viron, in Eagle’s minimum essential medium supplemented with 5% calf 1981; Fakan, 1994) and are associated with hnRNP core pro- serum. Twenty-four hours later, when cells were near confluence, they teins and spliceosome components (Fakan et al., 1984; Puvion were processed for electron microscope study. et al., 1984a; Visa et al., 1993a). This is consistent with the For the observation of unloosened nucleoproteins, cell cultures detection of snRNP antigens on transcripts which are still were fixed in situ, at 4°C, in 4% formaldehyde (Merck, Darmstadt, attached to the DNA matrix as seen in studies of spread tran- Germany) in 0.1 M Sörensen phosphate buffer, pH 7.3. During the 1 hour fixation, the cells were detached from the culture dish and cen- scription complexes (Fakan et al., 1986) and in observations trifuged. The pellets were dehydrated in increasing concentrations of indicating that splicing may occur co-transcriptionally (Beyer methanol and embedded in Lowicryl K4M (Chemische Werke Lowi, and Osheim, 1988; Wu et al., 1991). Waldkraiburg, Germany). Polymerization was carried out for 5 days Several years ago, we developed a new fixation procedure at −30°C under long wavelength UV light (Philips fluorescence tubes which helped to bridge the gap between the molecular organ- TL 6W) (Roth, 1989). Ultrathin sections were collected on ization of chromatin-RNP complexes in vitro after spreading Formvar/carbon-coated gold grids (200 mesh). of the nuclear contents, on the one hand, and the more compact For the observation of loosened nucleoproteins, culture medium was in situ structures observed in thin sections of routinely fixed removed and cell cultures were exposed for about 5 seconds to distilled cells, on the other hand (Puvion-Dutilleul and Puvion, 1980; water adjusted to pH 8.5 with 0.1 M borate buffer (Puvion-Dutilleul Puvion-Dutilleul, 1983). This method, adapted from Miller’s and Puvion, 1980). The cells were then covered for 1 hour with 6 ml of spreading technique (Miller and Beatty, 1969), consists of 1% formaldehyde aqueous solution containing 0.4% Photo flo (a com- mercial wetting agent containing Triton X-100; Kodak-Pathé, Chalon, preparation of cells in a detergent-containing hypotonic France), and 0.1 M sucrose. This solution was previously adjusted to fixative solution, which induces only a mild loosening of pH 8.5 with the borate buffer. During the 1 hour loosening treatment, nucleoproteins instead of the extensive dispersal of chromatin the cells were scraped and centrifuged. The resulting translucent pel- employed by Miller. Until now, our modification of Miller’s lets were dehydrated in methanol and embedded in Lowicryl as procedure has been used for studying in situ transcription described above for pellets of unloosened cells. complexes in normal cells (Puvion-Dutilleul and Puvion, 1980; Nuclear matrices of HeLa cells were kindly provided by F. Harper Vazquez-Nin et al., 1989, 1992), drug-treated cells (Puvion et (Villejuif, France). They were prepared from cells still attached to the al., 1981; Puvion-Dutilleul and Puvion, 1981) and virus- plastic culture dishes as previously described (Puvion et al., 1988) infected cells (Puvion-Dutilleul et al., 1980; Puvion-Dutilleul according to a slight modification of the method of Buckler-White et and Puvion, 1982), and for studying the insertion of viral al. (1980). Briefly, cultures were treated for 3 minutes at 4°C in TMS genomes into intranuclear viral capsids (Puvion-Dutilleul et buffer (0.25 M sucrose, 5 mM MgSO4, 50 mM Tris-HCl, pH 7.4) con- taining 1% Triton X-100 prior to being incubated for 30 minutes at al., 1989).